Severin Schwan - Chief Executive Officer and Director Daniel O'Day - Chief Executive Officer of Roche Molecular Diagnostics and President of Roche Molecular Diagnostics Roland Diggelmann - Chief Operating Officer of Roche Diagnostics Alan Hippe - Chief Financial and IT Officer

Sachin Jain - BofA Merrill Lynch, Research Division Richard Vosser - JP Morgan Chase & Co, Research Division Andrew C. Weiss - Bank Vontobel AG, Research Division Tim Anderson - Sanford C. Bernstein & Co., LLC., Research Division Michael Leuchten - Barclays Capital, Research Division Jo Walton - Crédit Suisse AG, Research Division Andrew S. Baum - Citigroup Inc, Research Division Odile Rundquist - Helvea SA, Research Division Peter Verdult - Morgan Stanley, Research Division Vincent Meunier - Exane BNP Paribas, Research Division Keyur Parekh - Goldman Sachs Group Inc., Research Division

Good afternoon, ladies and gentlemen. Welcome to our briefing on the Roche half year results. Overall, Roche delivered strong operating results for the first 6 months. You have seen this morning the financials. Sales are up by 5% on a group level, core earnings per share up by 12%. And in spite of the setback of Aleglitazar, we also made overall good progress on our pipeline. We moved Bcl-2 and anti-PDL1 into late-stage pivotal trials. There's a number of promising compounds coming through from early developments such as etrolizumab or anti-factor D. We further strengthened our HER2 portfolio with a positive CHMP recommendation for Herceptin subcu, so overall, a good momentum on the innovation front. On the sales side, Pharmaceuticals up by 6%, very much driven by our key cancer medicines, but we also saw very good growth of Actemra in rheumatoid arthritis, getting one of our key brands with a growth of 33%. Diagnostics up by 3%, an acceleration versus the first quarter, very much driven by the clinical Diagnostics business. So we are back to a good sales momentum in the mid-single digits, and it is very much driven by the U.S. in absolute terms, the biggest growth contributor, strong performance there. But we also see continued good development in our key emerging markets, 10% up in the first quarter, again 11% up in the second quarter, very much driven by the performance in China, up 26% in Pharma, 19% in Diagnostics. Turning to profitability, you can see that we bring the good development on the top line, also to the bottom line. Margins stands now at 41%. This includes a one-off impact from the past service income. But even if we correct for the past service income, margins would be up by approximately 1 percentage point at 39.4 percentages, and absolute operating profit growth would have been at 7%, outgrowing the sales growth. And that is also reflected in core earnings per share, up by 12%. If we would exclude the onetime effect of the past service income, it would be up by 8%. I mentioned that we moved Bcl-2 and anti-PDL1 into pivotal trials. We have now a rich late-stage pipeline with 10 compounds in pivotal trials. Let me just also emphasize again the excellent results we could communicate for GA1 -- GA101 yesterday, the head-to-head study against MabThera where we showed that we could significantly increase PFS in the direct comparison versus MabThera. So on the basis of those strong results, we confirm our outlook for the full year. Group sales to grow in line with previous year; core EPS to grow stronger than sales, and that excludes the PSI effect. And on the basis of that, we are confident to further increase the dividend. And with this, I hand over to Dan O'Day for Pharma. Daniel O'Day: So good afternoon from my side as well. I'm looking forward to updating you on the results of the pharmaceutical division for the first half of 2013. Let's dig a little deeper into the sales with the 6% increase in overall sales for the first half of the year. You can see we had a very strong first half in the United States. That's driven by a number of products, certainly the newly launched Perjeta and Kadcyla, but also Avastin, Actemra and Lucentis, amongst others, so a very good strong first half in the United States. In Europe, we grew at 1%, significantly ahead of the market here. In fact, when you look at the price differential or the price impact for Europe in the first half of the year, we anticipate that it's around 4%, which mean on volume terms, we grew at 5% in Europe, I think really speaking to the innovative nature of our products in that constrained health care environment. Japan growing at 2%, really quite good considering the fact that Evista, one of the products there, was returned to Eli Lilly, and with the strong growth of the oncology portfolio, we still see a growth on the top line, irregardless of the price decreases we also had in Japan last year. And then finally, international region growing at 5%, the E7 growing 11%, and I will get a little bit more into that in my talk. And of course, the strong sales line has led to a significant increase in core operating profit as well. To give you a couple of details on the P&L, we did have a significant impact on royalties for the first half of this year with Lucentis, Eylea, Humira and some one-off product disposals. As we go into the second half of this year for year-on-year effects, we expect that that will decrease over the second half of the year on a percentage basis. The cost of sales. Actually, when we look at cost of goods and period costs, we were actually growing below the sales growth. So the explanation for the slightly higher cost of sales is actually related to Tamiflu and Avastin royalties, which is obviously also generated from the very strong sales of those 2 products in the first half of the year. M&D, R&D in line, and G&I -- G&A, excuse me, predominately affected by the PSI effect, which Alan will talk a bit more about. So again, a 9% increase in core operating profit at a margin that's really at the top end of the industry. So when we look of the overall product progress in the first half of the year, I guess the first thing I would take away from this Slide 17 is that the innovative products, the new products, are clearly far outgrowing the products that we are -- or declining in our system. And as expected, Pegasys has declined as a result of the warehousing of patients at this stage, particularly in the developed markets. It continues to grow in the emerging markets. And here you can see the effect in Japan of the loss of Evista, again Japan overcoming this with a strong oncology growth. I'll talk a bit about the oncology programs in the next slide, but just to say that Tamiflu is clearly the result of the very strong flu season in the U.S., which occurred in the first 2 months of this year, and then really strong growth on the Actemra side with 32% increase in our overall sales there as well. So when we break down the oncology franchise, that's growing at 8% in the first half of the year, really, a contribution from across the portfolio, MabThera/Rituxan growing at 3%. This was affected also by the timing of some of the tenders and the phasing in the U.S., very good solid underlying demand with MabThera and Rituxan. When we look at the Herceptin franchise, that's grown at 11%, really solid demand across the 3 products that are now part of the entire franchise. Avastin at 12% and Tarceva making progress now with a 4% growth, good uptake in first line, and we're seeing stabilization in the second line as we expected. Now this is a nice picture to show what now 6 indications in Avastin are driving. Really, this growth is driven quarter-on-quarter by the ovarian cancer indication in Europe and then in colorectal cancer, the treatment beyond multiple line is really helping to grow in all of the regions around the world. So this is, again, a very strong picture for Avastin and really driving our overall oncology growth. Digging a bit further into Avastin with the exciting news at ASCO around the cervical cancer indication, we have the potential to add, in fact, a seventh indication for Avastin. We'll certainly be taking that study and approaching the regulatory authorities in the U.S. and Europe and having discussions with them about the ability to file that data for an indication in cervical cancer. And then also at ASCO, we had the news on the glioblastoma trials. We continue to feel confident that the data supports the filing in Europe. We will, of course, have discussions with the U.S., most likely in the second half of this year. And then just to remind you, in the United States, we have about 170 million of the total Avastin franchise in glioblastoma in the United States as we continue to look at that product in the United States market. So the HER2 franchise, 11% growth, really strong uptake of Kadcyla in the second line and beyond therapy areas in metastatic breast cancer in the United States. And very exciting, the TH3RESA study also read out, which is just yet one more trial that shows the efficacy of Kadcyla, in this case against physicians' choice in second line and beyond, so further solidifying the data on Kadcyla in that setting. Perjeta continues to roll out well in the United States with penetration increasing in the first-line setting by around 5%. We have seen there was some use in the second line and beyond setting off-label prior to the launch of Kadcyla. So we've seen Kadcyla obviously picking that business up, particularly in the United States. But the continued growth in the licensed indication in the first line continues. And of course, we're just beginning to roll out now in Europe with the approval there, and obviously, the time it takes for reimbursement. We've rolled out the product in Switzerland, in Germany and the U.K. with the vast majority of countries in Europe still in the reimbursement process. And of course, very exciting is the neo-adjuvant claim for Perjeta that we have filed in the United States, and the PDUFA date is October 31 for that as well. And then finally, the good news on Herceptin subcutaneous, again, a part of the entire franchise, we received a positive CHMP opinion on in June of this year. So we would expect to launch that in Europe and other markets outside the United States around the world in the second half of this year, which is again another strategy to protect the franchise against biosimilars as we move forward. So I have to say we're really encouraged by the data on Lucentis in the United States. I had been a bit more cautious at the beginning of this year around Lucentis. And I've been positively surprised to see us stabilizing our share really in AMD which, as you know, is the vast majority of the market. And not only that, but we see the entire market growing in AMD. We see some evolution from Avastin business into AMD into Lucentis and Eylea. Our share in RVO is stable. And of course, the growth for us continue to come also in DME, where we continue to penetrate that share-wise in the United States. So good strong growth for Lucentis. I have to say that we continue to see growth for Lucentis in the second half of this year as well. Now the E7 markets are growing nicely, 11% growth, pretty steady growth quarter-to-quarter overall, 26% growth in China, again, that's driven by the oncology franchise, good growth in Brazil and in other markets. We do have some tender effect, and I'll come back to this in Russia. There's been some timing of tenders that's affected the business in Russia, but the underlying demand is good and strong, and we expect that to resolve itself as we go into the second half. But it has had some impact on some of the demand for Herceptin and MabThera in our numbers at the first half. Now to dig a little bit deeper into the international region, which includes obviously the countries beyond the E7, here we really did have some more significant tender effects that are affecting this 5% growth, which is not truly representative of the underlying demand. You see the strong growth in China, again driven by predominantly the oncology portfolio, Turkey. But then you see in Algeria, which has the phasing of the tender is completely different from last year. So last year, we had it in the first half. This year, we expect it in the second half. So you get 2 effects there. And it causes a magnitude effect of about 131 million. But we do expect the business in Algeria to continue to be positive for us in the second half of the year, so I just wanted to explain a bit the tender effects there. So turning the attention to the innovation, well, Severin already mentioned this, the very first medicine ever that's shown superiority to MabThera. I mean, MabThera set a very high standard, and for the very first time in a head-to-head fashion, we've seen a product, and this product is GA101, be superior, significantly superior to MabThera in CLL, and this is very exciting data for us. I think it's very exciting data for the future of the hematology franchise, and we look forward to presenting this data later this year. We have submitted it to ASH, and we expect that you'll be able to see the more detailed data at that period of time. Again, I'll just remind you, GA101 in the United States has both breakthrough and priority review status, and we would expect to hear back on that by the end of the year. And of course, on the NHL programs, as we communicated at ASCO, due to the recruitment of those trials having been accelerated, we do have also the filing dates now for NHL and DBCL in 2015 and first line non-Hodgkin's Lymphoma in 2017. So PDL1, we're also very excited about. These are still early days I think for immunotherapy in cancer. We think we have good strong data on our anti-PDL1. It's encouraged us to move ahead now with a single-arm trial that's diagnostically enriched, if you like, an ability to look at the effect of this in non-small cell lung cancer patients that are diagnostically positive. And we have patients now ongoing in that trial, and we expect to have a readout of that trial in 2014. And we've also officially moved this into our late-stage development of PDL1. And we're looking now at metastatic non-small cell lung cancer also in second and third line in a more traditional Phase III trial. That Phase III planning is now underway, and we'd expect to be enrolling patients later this year or early next year on that with the potential filing for that in 2016. Obviously, the PDL program is broader than this. We're in the process of looking at combinations and other indications as well as we look at the promise of this new therapy. We made some exciting progress on the subcutaneous front. I've already spoken about Herceptin. I'll just remind you that for Herceptin, we're talking about x U.S. sales, which are about 65% of the sales and predominately in the adjuvant setting, which are around 70% as a protection mechanism and an ability to add a huge convenience to patients and to health care systems by administering infusion in minutes versus hours. We have the same strategy with MabThera subcutaneous. That is still under review, as we expect, in EMA. And I'll remind you in this particular case, we're talking about 50% of the sales outside the United States and the utility. This would really be on the oncology setting, both in the induction and the maintenance phase. Again, a stabilization strategy, a convenience to patients and ability to differentiate before biosimilars. And then really a growth opportunity is clearly in Actemra subcutaneous. You know the vast majority of the market is subcutaneous and biologics in rheumatoid arthritis. We're very pleased to say that in the U.S., this review is going well, in fact, so well, we've brought it forward now into the end of this year. And that opens up a significant portion of the biologics market, as you can see here, is 70%. So in the outlook side, just to get a little bit deeper into the portfolio that Severin presented, since we last spoke to you at the first quarter, we have officially moved the anti-PDL1 and the Bcl-2 inhibitor into late-stage trials. The Bcl-2 inhibitor, we're looking at p17 deletion trial, and we're also looking at it in combination with both MabThera and GA101 for the future, amongst other trials we're looking at for the Bcl-2. Now relative to the decision-making and the progression of the portfolio, we do now have data in-house, and we're looking at that data etrolizumab for ulcerative colitis and anti-factor D for the dry form of AMD. I would just say you've seen the data on etrolizumab that was already presented, and anti-factor D will be presented in August. So we're now analyzing this data and determining the right step forward, and we'll make a decision on both of these compounds by the end of the year. We have a number of other compounds in the middle section, which are yet to read out, yet to inform us, and this will be done in the second half of this year and the first half of next year. And then, we have made some decisions on a couple of other compounds in our portfolio that we'll look to partnering opportunities for. So we mentioned before the anti-PCSK9, we're in the process, for both competitive and intellectual property reasons, looking at partnering opportunities for that outside the group. We won't move ahead with that alone inside of Roche. And we've now received the full results from the ANNAPURNA trial, which basically was successful, but given the way the field has moved in HCV, we don't see this as being as competitive as other agents in the marketplace. So as you know, we have a focused strategy on danoprevir for China, and we'll be looking at mericitabine and setrobuvir to see if there are other potential partnering opportunities for those compounds as well. So the data you can expect in the second half of this year, as I said, anti-factor D, which will occur in Toronto with the end of August. At ESMO we'll give you some additional longitudinal data on the Phase I of the anti-PDL1 and the TH3RESA trial in Kadcyla. So that's coming up also at the end of August, early September. And then importantly also, the GA101 data, which will be presented -- well, we've submitted it and we hope it will presented, we assume it will be presented, in December in New Orleans at ASH. So just to conclude on the progress on the pipeline, we made very good progress, as you can see here. Of course, aleglitazar did not meet its expected endpoint, and we are now in the close-down phase of that. Of course, that's a disappointment to us and to patients. But having said that, it was a high-risk program. It's one that has quite a high bar to show significance in in cardiovascular disease. And in a way, it's of course, it's better to know sooner rather than later on a product like aleglitazar if it has some safety signals. So we're in the process of closing that down now. And of course, we have the Actemra subcutaneous in North America. We expect an indication on that in RA still in the second half of this year, U.S. approval, the neo-adjuvant in Perjeta and of course, the GA101 in front-line CLL as well. So a good overall quarter, not just on sales but also for the portfolio. So with that, I will turn it over to Roland to cover Diagnostics.

Well, thank you, Dan, and good afternoon from my side as well. It's a pleasure to be able to report on the Diagnostics sales and results for the first 6 months. As has been mentioned, 3% growth for the 6 months was good momentum in the second quarter, 4% growth in April to June month. And as you can see, the growth, very much due to strong growth in professional diagnostics, with 6%. Diabetes Care remains a challenge with minus 5%. As you may have heard, we have restructured the organization, integrated the Applied Science business, and the majority of sales are now under Molecular Diagnostics, which has had an impact on Molecular Diagnostics. The underlining growth for Molecular is a strong 4% and good growth as well in Tissue Diagnostics was 6%. So overall, the clinical laboratory business is doing very well with a 5% growth. Geographically, strong growth from emerging markets, both Latin America and Asia Pacific. Good growth also in EMEA, positive growth in a difficult market environment, same holds true for Japan. And also, North America impacted by Diabetes Care. But excluding Diabetes Care, positive growth of 2% also in the United States. Looking at the P&L, good momentum from growth translating to the bottom line with a 10% core operating profit increase, also due to good cost control. Some onetime effects that I would like to mention here that will not reoccur in the second half, mainly a favorable ruling on VAT in Germany to the tune of 30 million; and then a bad debt from 2012, which did not occur in the first half of 2013, in the magnitude of 47 million. So a combined close to 80 million onetime favorable effect on the core operating profit. In terms of the integration of applied science, we are reporting the numbers in a different structure. We have integrated qPCR biochemical reagents and sequencing into Molecular Diagnostics. We expect from this a better technology flow-through being able to better serve our customers and also drive internal synergies. Custom Biotech, which is largely reagents for pharmaceutical manufacturing, is now part of Professional Diagnostics, here too leveraging internal but also product synergies. So this is the new reporting structure, 4 business areas, Professional Diagnostics, Molecular Diagnostics and Tissue Diagnostics in the core laboratory setting, and then Diabetes Care. I'd like to share a couple of highlights in terms of technologies and products. In Professional Diagnostics, very strong growth in immuno diagnostics, 12% growth, again, double-digit growth in the very core fundamental business in immuno. At the same time, new products, which I will share with you at a later stage. Diabetes Care, I have mentioned that and I will talk to it a little bit more, a challenging environment, continues to be challenging. On the other hand, Molecular Diagnostics with the FDA approval of the EGFR test. HPV sales doing very well, namely in the United States with more than double sales, and also the FDA submission for primary screening of HPV testing, which we expect to go to a panel hearing. On Tissue Diagnostics, strong growth and continued penetration in EMEA and Asia Pacific and good growth and uptake in primary staining, which offsets some of the reimbursement and guideline changes in the United States. A few examples on the Professional Diagnostics, very excited. We have previewed our new fully integrated total lab automation system called the cobas 8100, which provides huge improvements in terms of automation, hands-off, high-volume throughput, a completely modular system that can be tailored to customer needs and really driving the efficiencies in the laboratory and for our customers. And on the other hand, 3 new tests that we have launched providing more in increased medical value, proGRP for the accurate diagnosis of lung cancer, and then 2 tests for the diagnosis and patient monitoring in immuno suppressors. Another exciting event is the acquisition of Constitution Medical, which will strengthen our commitment in hematology, provide us with a novel technology, also on the basis of increased automation in hematology, which is very much at the core of our business, at the core of the laboratory business, so exciting new technologies, novel and innovation. On Molecular Diagnostics, as I mentioned, the approval of the EGFR test by the FDA, which is a simultaneous approval for the label extension together with Tarceva. Very excited about this opportunity as it will run on the well-established, well-penetrated cobas 4800 instrument, and it complements our oncology panel. And then on Diabetes Care, it continues to be a challenging environment. Both reimbursement cuts this year, especially noteworthy, of course, in the United States with the decisions by CMS to reduce reimbursement, but also price pressure in Asia Pacific. We have and are in the process of continuing to restructure. We're addressing this in terms of reducing the cost base, integrating R&D, integrating or reducing the number of platforms, but then also addressing the commercial organizations in those markets that are affected by reimbursement cuts. On the other hand, we continue to invest in innovation. We continue to see also good uptake of differentiated, of innovative systems, notably the Accu-Chek Mobile, which is growing more than 40% year-to-date. But we also continue to invest in insulin delivery systems in continuous glucose monitoring. And according studies here will actually demonstrate that the benefit of structure tested, of integrated approach to diabetes management was helping better patient and better glycemic control. That brings me to the key launches of 2013. As you can see, we're well underway to deliver on our launch plan, 9 out of 13 and confident to achieve the last 4 launches within the planned period. And with that, I'd like to thank you and I'll hand over to Alan Hippe for finances.

Well, thanks, Roland. Thanks to the team for setting the stage. A couple of comments on finance. And you'd see here on this slide the finance highlights and with the headlines, the business highlights, the improved financial result. The net debt is down by $3.7 billion since June 2012. And then I have 2 rather technical topics I would like to talk about. I will touch upon all of these items in my presentation. When you look at the group performance, as Severin said, group core EPS growth was 12%. I think I don't have to elaborate about sales. That has been described well. But when you look at the core operating profit up 10%, let me mention here that the past service income effect, which I will elaborate about in a couple of slides, gave us, let's say, an uptick here. If you exclude this effect for the core operating profit, the growth would have been at 7%. Same applies to the core net income. Core net income up 12% in constant rates. If you exclude this effect, the growth would have been 8%. And on core EPS, 12% growth as outlined here on the slide. Excluding this effect, the growth would have been 8%. It is worth to mention, this effect has no impact on the operating free cash flow and on the free cash flow. So let me focus quickly on the operating free cash flow and especially on the free cash flow. When you look at the operating free cash flow, it goes up by roughly CHF 200 million. And when you look at the free cash flow, it even deteriorates a little bit compared to last year. So what's behind that? We had a major cash in last year from the Spanish government, so called Montoro plan. A lot of companies benefited from this. We got a cash inflow of CHF 730 million last year. And as you can see, on the operating free cash flow side, we were able to overcompensate this. So we had really a high bar. We overcompensated by CHF 200 million. What we could not overcompensate anymore on the free cash flow line was really the higher dividend by roughly CHF 400 million. With that, let me go to the P&L. And when you look at the royalties and the other operating income, and you know we're really dependent here on the success of our partners, so Humira, Eylea, Lucentis, good things to mention here, also the out-licensing income. And as Dan has mentioned already, it could well be that the momentum will slow down a little bit in the second half of the year. And why is that? Because we came in last year, and the second half was quite a good half. So the base effect from last year is quite a significant one. And you might remember in this time, we had Eylea milestone payment and also a couple of product disposals. But this is something which is a little bit volatile asset, highly dependent of success of our partners. Cost of sales. When you look at the cost of sales, 5% up, higher royalty expenses, good ones from MabThera, Tamiflu and Avastin, higher collaboration expenses especially in the U.S. and also positive the VAT refund Roland has talked about of roughly 30 million. When you look at M&D, M&D moving in the right direction, here also a benefit. Lower bad debt expenses compared to last year, which also shows a little bit that our receivables and credit management really goes in the right direction. And when you look at R&D, up 3.6%, majorly driven by late stage. I think here really the opportunities Dan has talked about really show, and 3.6% is well in line with everything we expect. G&A. In G&A, the PSI effect comes into play because you see a development of minus 20%, which at first looks very favorable. It is very favorable. The past service income effect of -- in the core operating profit, CHF 252 million, is booked here. So when you exclude this effect, G&A would have gone up by 7%. Here to mention, it's really the medical device tax in Diagnostics, as well as IT systems that we bring into the company. Core operating profit growth, 10%, excluding the PSI effect asset plus 7%. We look at the margins, and Severin has started with that. First of all, all margins went up, also in Pharma as well as in Diagnostics. The past service income has helped us here. But I can say very clearly, if you exclude the past service income effect on the group level in Pharma as well as in Diagnostics, all margins go up. And now let's come to the past service income, which is a very, very technical topic to deal with because it relates to changes that we have made in the group pension plans, and majorly in Switzerland as well as in the U.K. And these are very technical things. And for example, in the U.K., we have changed really the index, which demonstrates or brings in the future-related inflation assumption, if you like. And we went really from the Retail Price Index to the Consumer Price Index, which I perceive as being a very technical procedure. So it lends really a onetime IFRS accounting impact on the IFRS pension liabilities to these planned changes, which means this is a one-off reductions of provisions. Asset, no cash effect at all. We have not taken any cash from the pension plans, and it doesn't mean a reduction in company cash contributions. It might be that we have also a pretty small impact in the second half of this year coming from these actions. With that, I would like to go to the tax rate. And you see the tax rate increase from 22.7% to 23.1%, and the reason for that is a good one. Because the business runs well in the U.S., it runs very well in Germany, up 8%. It also runs well in France and -- but all these countries are high-tax countries. So well, unfortunately, the tax rate moved with it. My expectation for the tax rate at the end of the year is around 23%. When we look at the net financial results, well, it shows an improvement as hoped for and as predicted, so contribute to the core EPS growth. But let me do this really step by step. The first point to talk about is the currency impact, and the currency impact is majorly driven by the devaluation we have experienced in Venezuela. This accounts for CHF 62 million in this sheet here. When you look at the debt redemption, you might remember we have done a recall of USD 400 million for a bond due in 2019. That brought in here a minus CHF 78 million, and the minus CHF 32 million is just the difference to what we have done last year. So evidently, this year in the first half, we have done more than last year. And then you see the interest expense positively. It was CHF 139 million. And this is the result of our debt down payments we have done right at the beginning of the year. When you look at the operating free cash flow and the margin itself, you see really that for the group level, we're pretty much stable. We're also pretty much stable in Pharma, and we go down on the Diagnostics side. And here once again, worth to mention is the cash inflow that we have seen last year coming from Spain, which was very significant. So once again, on group levels, CHF 730 million cash inflow last year. We evidently overcompensated this. And for the Pharma side, that meant CHF 436 million; and for Diagnostics, it meant last year CHF 294 million, just to clarify the effect here. If you exclude this effect in the last year, the operating free cash flow would have gone up by 15%. The investments in PP&E and intangible assets are pretty much at the same level as half year 2012. Another rather technical point is the operating free cash flow and the impact of a reporting alignment that we have done, which in fact goes down to our equity compensation plans. We have refined the calculation of the free cash flow in 2013 to fully exclude the impact of employee stock options. And the reason for that is we have done a peer comparison, and we thought it might make sense to refine the methodology just to be in line with our peers. So what we have done is we brought all the cash outs and cash ins related to our equity compensation plans below the free cash flow line. And they are now shown as change in net debt, which is outlined in the half year financial report on Page 42. I've talked about the debt situation already, and that we have reduced our debt exposure quite significantly right at the beginning of the year. And you see it on the left-hand side on this slide. We paid down EUR 3.3 billion and we paid down USD 1.75 billion right at the beginning of the year, and it shows on the lower interest expenses. We have also done a call of USD 400 million of bond, U.S. dollar bond, which is due in 2019. Bottom line, 66% of the Genentech-related debt has been repaid. And that means of the CHF 48 billion we brought on to our balance sheet in line with this transaction, cumulatively, CHF 32 billion have been repaid at the end of June 2013. And that leads us to the net debt situation and also to the net debt to total asset ratio. Let me first talk about the net debt situation. It's a small number on the right-hand side of this chart. You see June 30, 2013, as a small number up there, the minus CHF 13.6 billion. This is the net debt status at the end of June. When you compare that with the year earlier, 30th of June, 2012, there, the number was CHF 17.3 billion. So we have a reduction of CHF 3.7 billion in a 1-year time frame. When you look then at the ratio, net debt to total asset, at the end of the year 2012, and this is the 16.4%. And when you really take the trajectory and the trend coming from this upper line, I think we feel and I feel comfortable that we will get into our targeted range of 0% to 15% net debt on total assets at the end of the year. Another point to mention, and this is a standard in my presentation, is the currency impact on the Swiss franc results in 2013. And you know how that goes. Assuming the June 28, 2013, exchange rates remain stable until the end of 2013, which is pretty unlikely, the 2013 impact is expected to be the following: And we look at half year with the minus 1% on sales, the minus 1% on core operating profits, the minus 2% on core EPS, this is very much driven by the Japanese yen. When you look at the full year, with minus 2% on sales, minus 2% on core operating profit and minus 3% on core EPS, the U.S. dollar comes into play a little bit. You see the impact has been at half year plus 1% from the U.S. dollar and will be at 0% at year end. And with that, let's go to the outlook once again, and I don't want to be redundant here. But very clear on one hand, the outlook is confirmed. But on the other hand, it is worth to mention that the one-off effect for the PSI is excluded when it comes to this guidance. And with that, thanks for your attention, and we're happy to take your questions.

[Operator Instructions] The first question is from Mr. Sachin Jain from Bank of America Merrill Lynch. Sachin Jain - BofA Merrill Lynch, Research Division: Sachin Jain from Bank of America. First the obligatory, I guess, M&A question for Severin. There's some comments on the wires this morning, so just your focused on bolt-on acquisitions, no particular size, dynamics given they're affecting those -- or any potential deals would be in the less than CHF 15 billion range or something like that. And then 2 product questions for Dan. Firstly on PDL1, you list the first study as completing in '14. Filing is listed as '16 and beyond. I understand that Bristol has a similar study with a response rate endpoint listed as potentially fileable. So I wonder if you could comment on what discussions you've had around filing on response rate, understanding that that doesn't seem to be your base case. And then final product question on the factor D. Just to clarify commentaries, is it fair to assume that the Phase II study has met its primary endpoint, given that you're discussing potential Phase III? I wonder if you could provide some color on the Phase II endpoint. I understand it's plaque, lesion progression. What would you view as clinically meaningful there?

Sachin, thank you very much. And may I start off with the M&A question. So indeed, there's no change in our M&A strategy. We continue to look for opportunities to bring external innovation into the company by licensing deals, by technology deals. Or as you have seen actually in the first half of this year by acquisitions like Constitution Medical. And we will continue with this kind of bolt-on acquisitions as we go forward. Dan, if you want to answer on PDL1 and factor D? Daniel O'Day: Sure. So Sachin, first of all on PDL1, I think you're right to point out that our diagnostically enriched trial is, I guess, similar to what we see the competition doing in those areas. Perhaps we're -- I think we're being appropriate. I don't know that we're being conservative. But were the competitors to actually be able to have a fileable product based upon their faster markets, single-arm trials and we also had a successful trial with our PDL1 diagnostically enriched, then it could be that that could also be a faster-market strategy for us as well. But having said that, we don't anticipate that in our base case, like you point out. We anticipate that in Phase III trial randomized will be needed here for FDA purposes. And therefore, that's why we have the expected first filing in 2016. So it will be in relation to the competition, I guess, is what I would say there. Anti-factor D, I'm afraid that I really can't respond to a lot of questions on that because the data is going to only be presented in Toronto at the end of -- it's a Phase II trial, as you know. So I would say that I really can't give you much until it's presented. Once it's publicly presented, I think we can discuss the data more robustly.

The next question is from Mr. Richard Vosser from JPMorgan. Richard Vosser - JP Morgan Chase & Co, Research Division: Richard Vosser from JPMorgan. A couple on the HER2 franchise, please. Could you -- you alluded to the penetration of Perjeta's gone up 5% in the course. Could you give us an idea of where current penetration is and also what sort of duration of treatment you're getting from Perjeta now it's been on the market for almost a year? And just on Herceptin at the same time, obviously, it's being eaten into by Kadcyla. Just how should we think of Herceptin from here? Should we think of it declining? Or should we think there's some increase in duration of use with Perjeta that helps balance this out? Just some idea there. Then on Lucentis, of course, an excellent result where the market is stabilizing. What do you think will happen once you see the AMD data -- or sorry, the DME data for Eylea? Do you think you will end up seeing more declines through '14 for the Lucentis franchise? And then just one question on R&D. With the aleglitazar termination and some products being chosen to out-license, could you give us an idea how you're thinking of reallocating or reducing the R&D spend because of those decisions going forward in the second half of '13 and '14?

Okay. Thank you, Richard. Perhaps I'll start off with the R&D question. Now it is correct that on the one hand, we have savings because we don't continue with the trials. But on the other hand, we have obligations to terminate all these trials. And there won't be a material impact for the second half. So the net effect of that will not be material. As far as next year is concerned, I mean, aleglitazar, of course, is one of the opportunities in our rich pipeline and a lot will depend on how the various compounds move on in our pipeline. You've seen we expect many readouts to come for the next months. And depending on the data we will see, this will, of course, also very much determine our investments into late-stage R&D. So we will give you a better indication at the end of the year when we give the guidance for 2014. For the HER2 franchise, Dan, if you could take that? Daniel O'Day: Sure. So thanks for the questions. The first 2, I'll answer around Perjeta and then the second one, Lucentis. I think maybe just to frame the questions around Perjeta, I do think more and more will be important, and I'll explain why that we look at the Herceptin -- the HER2 franchise, excuse me, combined. Because there will be a lot of interplay over the next months on this business, and I'll try to explain that. First of all, specifically to your question, how is Perjeta doing in the front-line setting? You're right. From our data, we see it's growing quarter-on-quarter, quarter 1 to quarter 2, about 5%, which brings this close to the overall 40% penetration in the first-line segment. Now I'd remind you, I mean, the current label for Perjeta is connected with docetaxel, which in the United States, of course, there is quite a bit of paclitaxel use. We do have Phase IV trials ongoing right now with paclitaxel, which could broaden, if you like, the comfort level of the chemotherapy backbone for that over the months and years ahead. So I think in terms of duration of treatment, yes, it is extending, of course, because of the extended survival. But I don't have the exact figures on that. The other thing that I would just remind you of when we take a look at Perjeta, in addition to the backbone therapy continuing to expand over time, is of course the new adjuvant, which we'll have towards the end of this year. And of course, the other big event for Perjeta will, from a data perspective, will be the MARIANNE trial, which reads out towards the end of next year at this stage. So we look at how Perjeta interplays with Kadcyla in the first-line setting. So it's a complex, interesting story. The good news is that there's several different ways it could play out over the course of this data. I'll also remind you that Perjeta is just beginning to get launched in Europe and in other markets in the world. So Perjeta is really progressing as we expect in the appropriate indication first-line setting. As I said, second and third line, we've seen some early displacement because of maybe some enthusiasm in the U.S. when it was first launched before Kadcyla came into that segment. And then on Lucentis. I think your specific question was relative to DME for Eylea. First of all, we don't expect that for a couple of years to really impact the marketplace there. Currently in the DME setting, we have probably about 1/4 of the market share. The rest of the market is Avastin and surgery options. But we see a nice continued uptake penetration of Lucentis in DME. And we think that will continue for the foreseeable future on Lucentis.

Thank you, Dan. In the meantime, we got in a question from the webcast audience from Herald Taas [ph] "When will bitopertin be available on the market?" Dan, correct me if I'm wrong, but filing is planned for 2014? Daniel O'Day: 2015 filing.

Filing for 2014 and launch for 2015. Daniel O'Day: Filing is '14. Thank you. You correct me, though.

Can we have the next question from telephone, please?

The next question is from Mr. Andrew Weiss from Bank Vontobel. Andrew C. Weiss - Bank Vontobel AG, Research Division: Could you give us an indication whether there's any kind of inventory buildup in Kadcyla? Or is most of that second quarter revenue growth is through underlying demand? With regard to the charge that you're taking for the hep C, can you give us a sense of what the remaining CHF 222 million are on the balance sheet, and what they actually represent? Third question, in the U.S. revenues for applied sciences are not disclosed anymore. But if we try to back it out, it looks like you're getting smoked in that market. Could you give us an indication of how big the impact is in the U.S.? And lastly, GA101, the -- it was a protocol interims analysis. Could you give us the boundaries for stopping that trial?

Okay. Thank you very much. Perhaps we start off with Kadcyla and GA101 with Dan. Daniel O'Day: Yes, of course. So for Kadcyla, no, we don't see -- and in fact, in the U.S. market, there really isn't an opportunity to do a great deal of inventory buildup because of the nature of how physicians are reimbursed there. So I think that really is true underlying demand. And again, I'd just remind you for Kadcyla, whereas Perjeta, the appropriate indication is first line. Of course, a cancer patient is only first line once in their life. Whereas with Kadcyla, you have the population of patients from second line and beyond. And I think that's certainly one of the explanations for the really strong uptake of Kadcyla. And I think the sales, if you see, really do represent demand. Secondly, on the HCV charge, yes, I mean, as you know, we have entered into an arrangement in danoprevir -- sorry, with danoprevir in China, and we're still considering some activities with mericitabine in particular. So that's the reason we hold some value on our balance sheet at present because there's value in those compounds as well. And then finally, I'm sorry, I do have to correct. But bitopertin is 2015 filing.

Okay. All right. It is 2015. Daniel O'Day: I just want to accelerate that at this stage.

Right. Could you also say something to GA101? What have been the boundaries for the trial? Daniel O'Day: Oh, I'm sorry. I didn't hear that question.

The last question, I believe, was what was the boundaries, specifications of the trial? But actually, we can't... Daniel O'Day: For the GA101 stage 2 reload [ph]

Right. Daniel O'Day: No, I'm sorry, we can't really comment on that. But we'll see more at ASH on the stage 2.

Okay. And Roland, if you could comment on the U.S. applied science business?

Yes, indeed. Applied science, we actually see some positive developments in qPCR and the nucleic acid purification cells. Also some margin positive from biochemical reagents, where we lose sales on the sequencing business, where we do play in the niche segment only, which is the long read, low-throughput segment. So this is the main impact on the sales. For the other businesses, actually, a solid performance.

The next question is from Mr. Tim Anderson from Sanford Bernstein. Tim Anderson - Sanford C. Bernstein & Co., LLC., Research Division: If I could go back to the earlier question on M&A, you'd mentioned still looking for bolt-ons. I tend to think of bolt-ons as maybe being in the single-digit-billions range. But I'm hoping that you can talk about what you see as the upper limit, generally speaking, for bolt-ons. And then a couple of pipeline questions. On GA101, do you -- can you remind us whether you think that the positive CLL11 results foretell us that in the NHL setting, you'll see a similarly big benefit? And because you saw this trial finish early, might we see the Phase III trials in NHL also finish early? Then on MetMAb, the drug's in Phase III for a big indication, which is lung, and you might have data out in the next 12 months or so. But it seems like it's been kind of quiet. There doesn't seem to be too much buzz on the compound. There wasn't very much of a presence at ASCO. And I'm wondering if you can just update us on your level of enthusiasm for this program and maybe firm up the time line for single lung data.

Thank you. So if I start off with the M&A question. This is really very opportunity driven. Whenever an opportunity comes along, we would look at it. And we would take a view on whether we can create value for our shareholders by seizing such an opportunity. So we are not looking at it in terms of specific values, but we really do focus on opportunities, where we build our existing businesses, be it products, be it technologies. And as you know, we are not in this big mega-merger acquisition. So I'm afraid I wouldn't be able to give you any specifics on potential opportunities which might come along in the future. With this, Dan, if you can take over for the product-related questions? Daniel O'Day: Sure. So thanks, Tim. For GA101, your question around positive CLL data portending NHL, I mean, I'd say a couple things on this. I mean, first of all, of course, MabThera first started with CLL and then went on to NHL. But I would be very careful. I mean, they're 2 different diseases, NHL and CLL. We'll have to see how they progress. And of course, we're very encouraged by the data in CLL and the magnitude of effect that we see. But I would never want to guess a clinical trial before we see the results. And likewise, I would stay away from trying to guess on whether an interim read in NHL would read out similar to the way the interim read read out in CLL. But having said that, let me just expand the discussion a little bit. Because of course, we're encouraged by GA101, but we're not putting all the eggs into that basket. So we've got the Bcl-2 inhibitor now that we're taking into late-stage development. We'll be looking at that in -- as a stand-alone agent in that p17 deletion refractory relapsed CLL. But we'll also be looking at it in combination with an anti-CD20 backbone, both MabThera and GA101. And the reason we're doing both is exactly because we're not exactly sure how the NHL studies will turn out. So we want to make sure we have many different eventualities covered. And I think we'll -- as we look at the ADCs, which as you saw in the slide that Severin and I presented, the antibody drug conjugates 22 and 79b, we'll also be looking at those in a similar fashion. Those combined with an anti-CD20 background, still to be determined whether that's MabThera, GA101 or both. But I think you'll see that in addition to single agent potentially superiority vis-à-vis MabThera and non-Hodgkin's lymphoma, we're also looking at a combination of therapies, both within Roche and potentially outside of Roche to create superiority to non-Hodgkin's lymphoma. So stay tuned. It's going to be an exciting period of time, but I think we have a lot of early encouraging data. And then finally on the MetMAb, yes, I'd love to be talking more about this. We're very excited about it. The Phase II results were extremely encouraging in this segment of patients in lung cancer. The second line non-small cell lung cancer trial is recruiting well. We expect a result from that in 2014, so that will read out in 2014. We're extremely excited about this compound. We're also looking at it in first-line non-small cell lung, both squamous and non-squamous. And we're also looking at utility in colorectal cancer in Phase II trials. So this is a compound because of the net expression in a variety of caners and the encouraging data we saw in lung in Phase II that we're excited about. I think part of the problem is we have so many products to talk about, we probably should talk about Met a little more frequently. But we're excited about it. We're looking forward to seeing data next year on this.

Thank you, Dan.

The next question is from Mr. Michael Leuchten from Barclays Capital. Michael Leuchten - Barclays Capital, Research Division: A couple questions, please. Number one, Dan, you mentioned the U.S. as a growth driver in H1. However, if I look at the quarterly progression, there are some soft spots in the U.S. for the oncology portfolio. I wondered whether you could shed some light on that. Secondly, on the comments you made about tenders, particularly the ones that are delayed, have you won any of those tenders since the quarter -- the half year has closed, to give us an idea how much of that might actually come sooner than later? Thirdly, just if I could go back to the R&D comment that Alan made, I thought you -- the commitment for R&D was to be flat. Now it's up, not much, but it's up in H1 in Pharma and also in Diagnostics. I just wondered how feasible a flat R&D budget really is given the portfolio that you have and the Phase II assets that are coming through and you have highlighted. And then lastly, just going back to MetMAb. I just wondered how the negative results in the triple-negative breast cancer fits into the comments that you just made in terms of the future settings for that compound.

Michael, thank you. If I just may give some more color on the R&D side, we have never said that it's flat. We always said R&D to be stable. And what we mean by stable is that we also account for inflation, for example, and certain fluctuations which come along the way, depending on what products are being moved forward into late-stage development. Well, we literally cap R&Ds for research and early development when it comes into late stage. We have fluctuations, and we adjust depending on the outcome of the respective trials. So in the first half of this year, it was 3.6%. This is substantially below the sales growth. You can argue that this is at the upper side of a stable development. But overall, I feel comfortable that we can continue with that pace as we go forward. Dan, on the additional question, if you could take that over? Daniel O'Day: Sure. So Avastin, as you see, is pretty strong in the U.S. in the first half. I'll comment specifically on Herceptin and Rituxan because we have had some phasing, I think, of the ordering and the sales pattern. But just to take Herceptin, I mean, you see that we had a 1% growth for Herceptin in the second quarter, but we had a 17% quarter-on-quarter growth in the first quarter of this year. I think it's more instructive to actually look at those 2 quarters combined. You get about a 9% growth. And therefore, you see that there is still good demand in the United States for Herceptin. Having said that, of course, with the launch of Perjeta and with Kadcyla, we would expect some erosion, of course, of Herceptin over time as those 2 products come into play. Now in Rituxan, it's kind of the same thing. In quarter 1, we had a 12%. In quarter 2, we had a minus 1%, again, due to some phasing of ordering and how that progresses. Overall, it's a plus 6% for Rituxan in the United States. And quarter 4 last year was 7%. So, yes, I don't think that's indicative of any type of demand, just to give you some hard numbers to that. So we don't see a slowness in oncology in the U.S. in any way. I think we see a natural evolution. And more and more, I'm going to turn your attention to the U.S., too, of looking at their HER2 franchise because there will be interplay between those 3 products. But then, of course, feel free to look at MabThera and Avastin separately. So let's look at the tenders delayed. Yes, I mean, I think some of these are delayed. So for instance, with Brazil, we have the tenders. So it's an issue of when it's ordered and how it's progressed, and the same thing in some products also for Russia. With Algeria, I think it comes with a bit of a rhythm. So we have confidence that we'll see the tender come through in the second half of the year. But to say that it's firmly awarded, I think, is a bit premature. But that process will continue. And then the final thing on triple-negative breast, I'm glad you brought that up. That's a trial that we started 2.5 years ago in our early research group Phase II. And at the end of the day, when we look at triple-negative breast, there is about 12% of patients in triple-negative breast, overall, that are Met positive. It's obviously a very, very difficult disease state to have anything working. And there were 3 arms to this Phase II trial. So when you broke down that 12% and when you broke it down into the 3 arms, it really wasn't possible to tease out on Met positivity rate that was at all meaningful in terms of number of patients to see any type of a trend. So I would discourage looking at the triple-negative breast cancer as a portend to future med trials. I'd much rather focus your attention on the Phase II lung, which had a sizable number of patients. We had looked at statistical significance with and without Met positive, and I think that's probably more indicative. But we'll see, of course, as the trials come out.

Thank you, Dan.

The next question is from Jo Walton from Crédit Suisse. Jo Walton - Crédit Suisse AG, Research Division: Three quick ones. You highlighted the tax rate going up because of the increasing proportion of profits in the U.S. Should we therefore assume that if it is 23% this year, maybe it's going to be a rising trend over the next few years? Still sticking on the U.S., I wondered if you could talk a little bit about pricing in the U.S. You've alluded to the net pricing effect in Europe. There's been some press speculation about your 340 hospital pricing and how discounts there are now $1 billion. So I wondered if you could just tell us a little bit about that. And finally, you referenced that one of the reasons for taking a partner on the PCSK9 was an intellectual property element. Does that mean that, effectively, you haven't got room to maneuver and we should think of this as only perhaps a less valuable asset for you?

Alan, if you can start on the tax rate, please?

I'll be happy to do that. And I think it's a very justified question. First of all, I have to say I think we give our guidance really on a yearly basis. So I think that's really the point. Look, I think the momentum at the moment is good. As I've indicated, for the full year, I expect really the tax rate to be around 23%. And I would even say I'm thinking for the next year, I think that's a range, what I feel comfortable with. Daniel O'Day: Sure. So U.S. pricing, 340 -- 340B, Jo, I mean, remember, this is something that went into effect last year or the year before that, I think. It's now been a couple of years. So it's really a part of our base business, I mean, just to describe what it is. I mean, it's a discount program that applies to certain hospital institutions in the United States that have a certain percentage of Medicaid-like patients or underinsured patients. And therefore, those hospitals -- the totality of the population in those hospitals, regardless of whether they're underserved or underinsured, receive this discounting. But this is not something revolutionary. It's something that's really been evolutionary in our business. In other words, it's -- year-on-year, we're seeing only a minor shift, I would say, from our business in the United States, our oncology business in the United States, to these 340B institutions. And although it may have a small effect over the course of the next several years, there is no major effect on that. And again, it's included already in the strong U.S. numbers that you see for oncology. And then finally, on the PCSK9, I'm afraid when it comes to IP issues, as you can imagine, it's really not productive to speak in detail about those. So I would leave it at that. And I would say that we're still in the process of looking for partners and looking for value for that asset.

Thank you, Dan.

The next question is from Mr. Andrew Baum from Citi. Andrew S. Baum - Citigroup Inc, Research Division: A couple of questions, please. Firstly, just back to your BD strategy, you obviously historically tried hard to diversify your portfolio away from your dominant presence in oncology with limited success. Could you give us some sense of your current focus, whether it's more in Pharma or in diagnostic asset? And separately, could you comment on how much value you see given the valuations of some of the U.S. star technology companies? And then the second question is, could you outline how far away from clinical development are your other checkpoint inhibitors and pCR-based approaches in oncology?

Okay. So let me start off with the business development strategy, how we look at developing our Pharma versus the Diagnostics business. It's really opportunistic. It is really a method of what opportunities are coming up. Can we create value around our core franchises? And we look both at Pharma and at Diagnostics. And then in terms of making a decision to proceed or not, it's many, many factors. Of course, the price is one factor. The strategic fit, obviously, is extremely important; potential synergies we can generate between a target and ourselves. Is there a cultural fit in terms of the people we take on board, et cetera, et cetera? So it's really -- I think what I'm saying is it's opportunity-driven and every case is different. And you have to look at all the criteria, which typically are very, very specific for the respective opportunities. That has always been the case, and we will proceed like this as we go forward. On the oncology side, Dan? Daniel O'Day: Yes. So, Andrew, I have to admit you kind of stumped me on this one. But I would look it up. So we do have check inhibitors, I know, in Phase I and dose escalation studies, but they're early. They're early, I would say that. And I would -- if I can offer to you, I'd offer that we get back to you with some more details on that.

Thank you, Dan.

The next question is from Odile Rundquist from Helvea. Odile Rundquist - Helvea SA, Research Division: So a few remaining questions. On Zelboraf, could you maybe give us a bit your outlook for the rest of the year and especially 2014 as we expect the entry of a new branded competitor or 2 new branded competition on the market? And when should we expect the Phase III readout of your combination of Zelboraf and your MEK inhibitor? Is it in the first half or second half of 2014? And then on Avastin, you showed very good growth due to a very intensive but also second line, I guess, colorectal cancer. Can you give us a bit more color on the market share you have compared to also competition, how you understand it?

Dan? Daniel O'Day: Yes. Okay, very good. So on Zelboraf, clearly, you saw the nice growth in the first half of this year. We're significantly penetrated, I would say, now in the United States with a high testing rate and a high treatment rate. And of course, we now have the competition that has approval for the 2 additional agents still awaiting approval for the combination of the 2 agents. But of course, when that combination comes through, we would expect competition in the U.S. market. And whenever -- if that gets approved outside the U.S., then we would also expect that type of competition going into 2014. So I would say that we'll see some dynamics, of course, in Zelboraf with competition in 2014 on that. Relative to our own trial that looks that our own MEK inhibitor and Zelboraf, we expect that data to read out next year in 2014. And then we'll -- obviously, depending on the outcome of that file, then progress that ahead. For ovarian cancer penetration rates -- sorry, I'm just looking at -- okay. So I'm just -- it just wasn't at the top of my head. It's above 30%, I would say, the ovarian cancer penetration rate for Avastin in the EU market.

Thank you.

The next question is from Mr. Peter Verdult from Morgan Stanley. Peter Verdult - Morgan Stanley, Research Division: Peter Verdult here from Morgan Stanley. Severin, just on strategy, same question, asked a bit differently. If we look at the pipeline and how that's developing, the outlook for oncology is looking very positive. The outlook for cardiovascular metabolic and the hep C franchise is clearly less positive. With the central nervous system asset, it's high risk. And the outlook is slightly uncertain depending on how those studies roll out. So I suppose the question to you is how comfortable are you with Roche becoming less diversified from a therapeutic area perspective going forward, especially given all the changes in dynamics we're seeing in the field of oncology? So that's the first question to you, Severin. Dan, just quickly on the anti-PCSK9. IP aside, this was -- and I think you were talking up at the R&D Day last September. I just want to better understand what's actually changed since September to make you seek a partner for this product. Were there -- I wasn't aware of any incremental data. And then lastly, Alan, sort of a repetitive question for me. But we're hurtling towards this 15% net debt to asset leverage target by the end of the year. Could you just refresh us on how you're thinking about balance sheet utilization over the next 18 months?

Thank you, Peter. The question of diversification beyond oncology, I think the answer really is we follow where the science takes us. We are not thinking in terms of therapeutic areas. We are thinking about it in terms of, do we understand the biology? Do we understand the science behind it? Do we think we have the capabilities, the technologies, to make a difference in those respective diseases? And from that perspective, I'm absolutely sure, over time, there will be scientific progress and new opportunities coming up also outside of oncology. Oncology, of course, will remain important for us. But I do believe that we should not focus in -- only on oncology but that we should be open for opportunities outside. Now having said that, the risk profile indeed is different, but so are the opportunities. If you take CNS for example, of course, if you have a Phase II trial, say, in Alzheimer, you can measure whether the plaque is removed, but you cannot measure any improvement in cognitive -- typically, you cannot measure any improvement in cognitive function. So that means entering larger trials mean that you take much more risk. On the other hand, of course, if you are successful, then it's a huge opportunity. And I think at the end of the day, with the portfolio, it comes back to the right mix. So you take bets which are less risky, perhaps less opportunity. But I think a company like ours has always done this. And I think we should continue to do that to also take some higher-risk opportunities with the potential of very high returns. There were questions for -- Alan, if you start off on the balance sheet side?

Yes, well, a question about the target range of 0% to 15% net debt on total assets and what we're going to do, yes, if you really achieve that range. Yes, first of all -- let me say first things. First, let's get into the range. I said, I think, it's pretty likely that we can achieve that this year if everything goes as it has gone so far. And yes -- but the other question for me is in the major one. When we drop out here, there is still time left. And when, I think, we will make up our mind and look at this and once we have a concrete intention plan idea, I think we're happy to address it. But for the time being, the target is to get within that range. And I think then we -- when we have the risk to drop out, I think then we'll come back.

Right. And we shouldn't have a technical problem to deal with that. I think it's actually a good problem to have. Then there was a question on the Pharma side. Daniel O'Day: Sure. Just to clarify now, the ovarian cancer market share is around 36%, 35% to 36%, just so you know. Now on the anti-PCSK9, you're right. We were more bullish on -- at the Investor Day or the R&D Day last year. I would say that the product continued to perform relative to how we expected, possibility for decreased duration of dosing. But what did change since then was the IP landscape. Things did adjust between when we spoke to you last September and when we eventually made a decision this coming spring. So I don't think there is any one event that caused us to change our opinion about internal development versus partnering. It was a series of events. And it just got to a risk level where we determined that this is a product that would not be in our best interest, given all the other opportunities we have in our portfolio to proceed with. And that's when we decided to make the decision to go out and seek a partner for it.

Thank you, Dan.

The next question is from Mr. Vincent Meunier from Exane BNP Paribas. Vincent Meunier - Exane BNP Paribas, Research Division: First, a follow-up on the performance in emerging markets. In Pharma, it was plus 2% at constant exchange rate. And you were talking about tenders in Brazil and in Algeria. Is this the unique reason for this relatively soft performance? The second question is on GA101 and the timing of development in lymphoma. Because in CLL, over the past few months, indeed, the program has accelerated. And you have completed the trials well in advance. Is it fair to assume the results in lymphoma could be published before '14 in refractory indolent NHL and in '16 for the first-line NHL? And the last question is an update on the search for a new chairman. Is there any possibility to come back to the former system of the combination of the 2 positions, Chairman plus CEO?

Okay. So let me just start right away with your last question. No, there is no intention to combine the chairmanship and the CEO role. I feel very happy as CEO. So that means we will get a new chairman. And as previously announced, we should be able to communicate that still in autumn this year. Emerging markets, Dan? Daniel O'Day: Right. So in emerging markets, if I understand your question, I mean, we had the 5% growth in the emerging markets. You referred to a 2%... Vincent Meunier - Exane BNP Paribas, Research Division: It was in Pharma -- in Pharmaceuticals, in international, in the press release, it's 2% at constant exchange rate. Daniel O'Day: Okay, quarter 2, I'm sorry. You're speaking to quarter 2. I'm with you now. Okay, yes, quarter 2. Yes, but back to your question, I mean, is this predominately explained by tenders? The answer is yes. I mean, these are really -- 130 million in the emerging markets is a significant amount that can fluctuate the growth rates. So as I said before, we have some delayed tenders in Brazil but not a lack of demand there. We've got Russia really going through a change in the way they are tendering on an annual basis, which is causing the impact here. And then we have Algeria. And we have Iran, which is about another 30 million or so. So when you add all of those up actually, it really is explained vis-à-vis the lower quarter 2. And I think when you see the quarter 3 numbers, you'll see some different figures in that direction as well. And then the second question on GA101, yes, I appreciate your optimism on that. I'd love to be as optimistic as you. But I have to say, I think, in terms of the speed of the trials, we're still expecting that -- assuming the trials go to their endpoint, which is our assumption, that the 2015 would be the filing time line for DB ACL [ph] and refractory and then 2017 for NHL. So I can't give you different dates on that. Of course, if we were to get a more aggressive data, we'll let you know quickly and act on it. But those are really the dates that we're working by, and we think they're the most realistic dates.

Thank you, Dan.

The last question for today is from Mr. Keyur Parekh from Goldman Sachs. Keyur Parekh - Goldman Sachs Group Inc., Research Division: I have 3, if I may. First, Dan, if you can help us think about the opportunity for Avastin in cervical cancer, where we saw great data at ASCO, just wondering how you guys are thinking about monetizing that opportunity. Secondly, Alan, if you can talk a bit about kind of the margin progression for the first half, obviously, Pharma kind of doing very well, Diagnostics improving. But I just would love to get your sense for where you see you are from a restructuring perspective in the Diagnostics side and how much opportunity there is to get more. And thirdly, kind of just in sense of -- I guess, your sense of where do you see the greatest opportunity amongst the late-stage assets? Obviously, MetMAb, you're really excited about. But as things are progressing forward, anything that you think might be meaningfully underappreciated by us?

Dan, do you want to start off? Daniel O'Day: Sure, sure.

We need another hour for promising opportunities in late-stage, but... Daniel O'Day: I'll do my best.

I'm listening how you prioritize. Daniel O'Day: So let me start with cervical cancer. Cervical cancer is a really exciting opportunity. You also saw it at ASCO in terms of the opportunity in a country like India in terms of being able to identify it early. I mean, the reality with cervical cancer disease is that the vast majority of the disease is in the developing world or in the emerging world. In fact, the incidence of cervical cancer in Brazil alone is bigger than the United States and the EU5 combined, just to give you a magnitude of that. So there is significant opportunity here. I think the question first and foremost is, can we get approval for the indication based upon the trial? And that's where we're really focusing on now and speaking with regulatory authority. So we'll have to come back to you after those discussions and inform you further whether an approval is possible in that indication. And then secondly, it really becomes an access issue. And again, you know we've had creative ways of approaching public markets in Brazil, for instance, with MabThera, where we have differential pricing in the private and public market and in other markets around the world. So this is something that, again, we would have to then be creative about how we find access for Avastin into the markets where cervical cancer has its highest incident. So of course, we weren't expecting this data at ASCO. So we're just starting to get our plans around that. And then relative to the portfolio, I mean, I would say we've talked about MetMAb. We've talked about the MEK inhibitor with BRAF coming out here as well, GA101, of course. I would say the other thing that I'm excited about in the field of immunology outside of cancer is lebrikizumab. So we're now beginning the full Phase III trial of lebrikizumab. We had the second Phase II trial that really reconfirmed the results from the initial Phase II trial that showed pretty significant reduction in exacerbations in a diagnostically focused population with periostin. So I mean, I think that's something that's quite interesting. And I mean, I think also, the CNS portfolio is something that we're going to know next year on bitopertin and schizophrenia, exactly how that stands out. Ocrelizumab, we know, has effect in MS. The question will be the robustness of the safety profile as we get into larger patient populations. And gantenerumab is just the first of their kind with all the risk that goes along with that but all the potential. So I would say those on the late-stage side, that's where we're at. And on the mid-stage or the Phase II, I think we've highlighted the ones we have data on. PI3 kinase is something else that's going to be coming through here soon that we'll take a look at the data on the antibody drug conjugates. So I think we'll be able to inform you more and more as the months go on in the second half for this year and the first half of next. I'm interested. Severin, would you have picked different compounds?

Yes, I would actually agree with you. I had also lebrikizumab on the top of my list. So we are very much aligned here. Thank you very much, Dan. Alan, can you give us flavor for the second half?

That's exactly the point, Severin. I think you framed it well here. Keyur, as you can imagine, I can just talk here directional. First of all, I'm very happy with the margin progression that we have made in the first half. And let me outline once again that the margin improvement was certainly driven as well by the past service income. So take that out. I think all margins have improved, including Diagnostics. So I would say things are really progressing well. We are really taking advantage of the leverage we're having. So I think that moves fine. I also expect to move on like that in the second half. And you know in the second half, traditionally, our margin goes a little bit down compared to what we have at half year. But I think really, at full year basis, we're heading well. My point is a little bit as Roland pointed out with his margin. I think he had a couple of one-offs in the first half. And I think you should respect that. We have given, how should I say, you guys the feeling. I think stabilizing the margin in Diagnostics would be already an achievement. My takeaway in the first half -- and I'm far away from judging really Roland's work, but I think that moved well. Look at the cost development that you see in Diagnostics, which is a quite favorable one. And certainly, this is also driven by all the restructuring he has implemented.

Thank you, Alan. And on that positive note, I'd like to thank you for your interest and close our briefing. Thank you very much.